This project proposes multiple approaches to the general question of the role of recombination in repair, mutagenesis and genetic adaptation of bacteria. Recent results for provided evidence that the phenomenon know as "adaptive mutation" (Cairns) can be explained purely by standard genetic events that occur during selective growth of cells carrying an amplification of the mutational target. The Amplification model proposes a sequence of genetic events occurring within the developing revertant clone - amplification - mutation - segregation - haploid overgrowth. It argues against the widely circulated ideas of stress-regulated mutation (directed or general), stationary phase mutagenesis or mutagenic recombination. We will continue to characterize this system and test detailed hypotheses to explain the remaining questions, which include the following. What are the exact mechanisms by which amplification induces SOS and activates DinB-dependent rnutagenesis? Why does selection-stimulated reversion require that the gene under selection (lac) be located on an F'plasmid? Why does non-essential, reversion-associated general mutagenesis require that lac be located on the specific plasmid (F'128)? Why is growth of the amplification clone inhibited after appearance of the lac reversion? Why do some amplification clones reach maturity (full-sized) revertants without ever achieving lac reversion? We will experimentally measure rates for each step proposed by the model try to mathematically describe the process by which the Cairns system completes a long series of genetic steps within on week of growth under selection. The model promises to shed light on the process of gene evolution, adaptation of pathogens to their hosts and multi-mutational origins of cancer. While mechanisms of replication, recombination, and repair are understood in great detail, it is less clear how these systems interact and how they achieve the amazingly low rate of mutation seen in bacteria. We will continue to pursue a set of assays that measure aspects of recombination as it occurs within the chromosome of growing bacteria. These assays do not involve crosses and require endogenous sources of strand ends to initiate recombinational repair. Thus they can be used to identify endogenous metabolic sources of DNA damage and the role of long-range replication in completion of a recombination event.